Previous studies of visual memory in ventral temporal lobe cortex have been performed by presenting images to monkeys whose gaze is fixed. Preliminary evidence from Ringo's laboratory has shown that over one-third of the units recorded in memory related areas the temporal lobe, (inferotemporal cortex, parahippocampal gyrus and the hippocampal formation) show significant saccadic modulation. This modulation is apparently not a consequence of a shift in visual scene because: (1) cells show saccadic modulation in complete darkness, (2) many cells modulate before the saccades, and (3) non-visual cells can show saccadic modulation. This has led the investigator to suggest that widespread coherent modulation of temporal lobe neurons by saccades may be important in facilitating memory. The experiments described employ single unit recording in behaving monkeys to test the four hypotheses. The first hypothesis is that saccadic modulation is state dependent. That is, it is augmented by general arousal, selective attention to vision and/or purposive command. This will be tested by recording saccades in a broad area of the temporal lobe during auditory and visual tracking and discrimination tasks, and at rest. The second hypothesis, that saccadic modulation has a subcortical route, will be tested by using split brain, cut chiasm monkeys. A subcortical route, will be likely if light delivered to the contralateral eye is capable of modifying the discharge of cells that saccade-modulate in the light in the ipsilateral hemisphere. The third hypothesis is that saccadic modulation has a non-visual role and is therefore found in non-visual cells. To find a population of non- visual cells, studies will be expanded into the upper bank of the superior temporal sulcus, where polymodal cells can be found. The fourth hypothesis is that saccadic modulation functions as a memory modulator and increases single unit memory effects. This will be tested by comparing the amplitude of the decrement with repetition phenomenon in a saccade task and in a fixed gaze task. The prediction here is that decrement in response to a reyeated presentation of a novel stimuli will be greater when the monkey uses saccades to view the new images compared to fixed gaze viewing (the greater the decrement the less novel or the more "remembered" is the cue).